S4E25 Cluster Analysis
Quantitude
00:00
How to Identify Optimal Clusters for Your Data
K means clustering. Well K is the number of clusters and means indicate that we're trying to compute the means within cluster. What we've done is unabashedly without guilt data driven to the core. We do not say, oh, consistent with theory, I have three clusters. This is, oh my gosh, I have 200 kids, 20 features. And I'm just trying to figure out organizing structure with a smaller dimensionality than helps me understand the complexity of this. That at its core is what we're doing here.
Transcript
Play full episode
